sign in sign up
<<
Home , Book of Documents, Teng (mythology), Yi Zhou Shu, Yu Gong

饒宗頤國學院院刊 第五期 417 2018 年 5 月 頁 417– 445

Origins of Chinese : Studies in the Composition and Thought of the Shangshu (Classic of Documents). Edited by Martin KERN and Dirk MEYER. Leiden and Boston: Brill, 2017. Pp. vi+508.

Edward L. SHAUGHNESSY East Asian Languages and Civilizations, The University of Chicago

The publication of a major English-language book on the Shang 尚書 (Elevated documents) or Shu jing 書經 (Classic of documents), the second of the , should surely count as a major milestone in the Western study of early . As the editors note in their Introduction, the Shang shu has inspired all aspects of Chinese political philosophy for over two thousand years now. Yet, as they also say, “In some kind of reverse—and bizarre— Book Reviews correlation, the Shangshu is as important to the Chinese political tradition as it is neglected in Western scholarship” (p. 2). Their claim just above this that “major Western works on the Shangshu can be counted on two hands, with fingers to spare” is only a bit exaggerated.1 In this volume we now have fourteen studies in just over 500 pages, that directly address at least fourteen different chapters of the Shang shu, not to mention two chapters of the

1 True, I count only eight or nine such studies listed in the various bibliographies attached at the end of each chapter, but they do not even include mention of such classic studies as Paul Pelliot, “Le Chou King en caractères anciens et le Chang Chou che wen,” Mémoires concernant l’Asie Orientale 2 (1916): 123–77, or Benjamin Elman, “Philosophy (I-) versus Philology (K’ao-cheng): The Jen-hsin tao-hsin Debate,” T’oung Pao 2nd ser. 69.4–5 (1983):175–222, or even Michael Nylan, “The Many Dukes of Zhou in Early Sources,” in Statecraft and Classical Learning: The Rituals of Zhou in East Asian History, eds. Benjamin A. Elman and Martin Kern (Leiden: Brill, 2010), 94–128, a study of the ways that the 周公 is represented in the Shang shu and the and which was published in a book edited by one of the editors of the book under review. Still, the point is well taken. 418 419

Yi Zhou shu 逸周書 (Leftover Zhou documents).2 The volume is the product Apparently Brill was more lenient in terms of page number than are most of two international conferences, one held at Princeton in 2013 and the presses, two other chapters in the book also being more or less similar re- second at Oxford in 2014, the published papers revealing considerable publications of studies published within the last three years. However, since revision and strong editorial hands. The fourteen chapters also display the the book in which the earlier version of Kern’s essay was published is probably various contributors’ different strengths, some engaging in deep reading of no more readily available than the present book and as far as I can tell has not the text(s) in question, others soaring over several texts or even whole books. yet been reviewed, the essay deserves introduction here. Different readers will come to different evaluations of the relative strengths Kern makes two principal arguments regarding the “Yao dian”: first that and weaknesses of these offerings, but taken as a whole the volume surely the text opens with a “performative speech,” and second that the overall effect makes the significant contribution promised in the editors’ Introduction. of the argument was understood in the and —and perhaps was created Indeed, to my mind, the only significant failing of the book is the Introduction in the Qin and Han—as an argument in favor of a particular view of kingship. itself, which is marred by an unappealing self-congratulatory triumphalism, He begins with a lengthy revisionist reading of the first sentences of the which simultaneously denigrates past scholarship—always without explicit chapter (here presented without punctuation so as not to prejudice the reading): attribution, while also suggesting that all of its contributors speak in a single voice, which is most certainly not the case. In the following remarks, I will 曰若稽古帝堯曰放勳欽明文思安安允恭克讓光被四表格于上下 first review the contents of the fourteen chapters, which after all constitute the heart of the book, before turning to consider the Introduction itself. This is usually translated as something like: ***

Examining into antiquity, Di Yao was called Fangxun. Respecting bright Book Reviews virtue and thinking peacefully, truly respectful he was able to yield, his Martin Kern is the author of the first contribution to the book: “Language and radiance covered the four exteriors and he caused those above and below the Ideology of Kingship in the ‘Canon of Yao,’” a study of the “Yao dian” 堯典 . to arrive. This is a revised version of a study by the same title published just two years earlier.3

Kern argues instead that “Fangxun” 放勳 , traditionally said to be the name of 2 The chapters are (in the order which they appear in the book, with the chapter number and the 堯 author of that chapter): “Yao dian” 堯典 (Canon of Yao; 1 Kern and 2 Vogelsang), “Gaoyao Yao , should begin a speech by Yao, and mean something like “imitating [past] mo“ 皋陶謨 (Counsel of Gaoyao; 2 Vogelsang), “Lü xing” 呂刑 (Punishments of Lü; 2 merits.” In offering this revisionist reading, he admits that he flies in the face of Vogelsang and 13 Sanft), “Gu ming” 顧命 (Testamentary charge; 3 Meyer), “Kang Wang zhi gao” all early readings, which he would date to the Han, as well as explicit evidence 康王之誥 (King Kang’s announcement; 3 Meyer), “Duo shi” 多士 (Many sires; 4 Gentz), “Duo in two different passages of the Mengzi 孟子 that Fangxun was Yao’s name. fang” 多方 (Many regions; 4 Gentz), “Jin ” 金縢 (Metal-bound coffer; 5 Gren and 6 Meyer), “Gan shi” 甘誓 (Harangue at Gan; 8 Kern), “Tang shi” 湯誓 (Harangue of Tang; 8 Kern), “Mu In rejecting fangxun as Yao’s designation, I consider the readings given in shi” 牧誓 (Harangue at Mu; 8 Kern), “ yi” 無逸 (Without ease; 10 Pines and 11 Hunter), “Bi shi” 費誓 (Harangue at Bi; 12 Khayutina), and “” 禹貢 (Tribute of Yu; 14 McNeal), as Shiji, Da Dai liji, and other Han sources to be misinterpretations. At the well as Yi zhou shu “Shang shi” 商誓 (Harangue at Shang; 4 Gentz) and “Wang hui” 王會 (Royal same time, the fact that already the Mengzi understands fangxun as Yao’s convocation; 14 McNeal). personal name raises two possibilities: either this reading, which runs 3 Martin Kern, “Language and the Ideology of Kingship in the ‘Canon of Yao,’” in Ideology of against the structure of the Shangshu text itself, was indeed very early, 香港浸會大學饒宗頤國學院Power and Power of Ideology, eds. Yuri Pines, Paul R. Goldin, and Martin Kern (Leiden: Brill, possibly in a separate tradition of the Yao legend, or the two pertinent 2015), 118–151. It is unclear whether such prompt republication is meant to signal the essay’s Mengzi passages (5.4 and 9.4) were composed only under the influence of unique importance or its urgent need for revision. Kern says that this “version now supersedes the earlier one” (p. 23 * note). However, a comparison of the first ten pages of the two versions Han sources such as the Shiji (p. 28 n. 21). shows them to be essentially identical—other than formatting changes—with the exception of an added paragraph on page 29 of the book under review, and deleted paragraphs at pages 125 It is not at all clear what Kern gains from denying all of this evidence. After and 126–27 of the previous version (the latter of which, at least, concerns primarily the Shi all, other than the first two sentences, his punctuation, and indeed his jing 詩經 [] and thus would be out of place in the present book). 420 421

understanding, of the rest of the passage in question is not different from the of the part. He concludes this analysis by saying “This suggests that in the traditional reading. The price for his suggestion to read this as a “performance Shangshu we must reckon with chapters that are neither homogeneous units text,” a familiar theme in his scholarship, is very high: the discounting of nor composite writings where the different layers complement or reinforce one virtually all counter evidence. A less interested reading might find that price another, but that actually contain rival discourses which may ultimately be too high. incompatible” (p. 78). Kern’s distinction between an activist Yao who regularly rejects the After a similar parallel analysis of the “Yao dian,” finding “two distinct advice of his advisers and a more compliant Shun 舜 who is guided by his parts that differ in vocabulary and political theory,” one “centered on the virtue advisers is interesting and well taken. He finds corroboration of this difference of the ruler” and one “mak[ing] a case for bureaucratic government in which in the “Guwen” 古文 (Ancient text) Shang shu’s division of the chapter into ministers are the most important actors” (p. 88), Vogelsang then moves on to two separate chapters: “Yao dian” and “Shun dian” 舜典 . When he says “Yao the “Lü xing” chapter. Noting that although it has not been split into separate and Shun could have been understood as representing two complementary parts in traditional editions of the Shang shu, he again divides it into two aspects of imperial rule that could be alternately actualized according to the separate parts, A and B. However, while his part A has seven separate sub- situation” (p. 54), this too rings true. His notice that Wang Mang 王莽 likened parts, B is represented by just one passage, which is a “catalogue of ‘procedural himself to Shun when he assumed power, enshrining Shun in the imperial law’ and other legal rules.” Whereas part A focuses on “virtue,” the passage temple as his ancestor, is likewise well taken. But when he goes on to imply— identified as part B emphasizes “penalties” (fa 罰 ), a word which Vogelsang without ever really stating—that this owes to Qin and Han political views, notes occurs seventeen times in the passage (without occurring a single time apparently disregarding the possibility that such views may have been debated in what he identifies as part A). He proposes two different scenarios to explain

in earlier times, he again seems to go well beyond where his evidence would this sort of difference: “one may argue that different contexts (e.g., the ‘king’ Book Reviews take him. After all, the tension between the prerogatives of the sovereign and speaking to different groups) or literary genres (e.g., speeches vs. lists or those of his advisers was perhaps the preeminent debate in Chinese political catalogs) require different styles and vocabulary and that the divergence in philosophy, important in the Qin and Han, to be sure, but also just as important content between A2 and B1 may be understood as supplementary rather than in earlier (and later) times. contradictory. Or one may take these discrepancies seriously and argue that In his contribution, “Competing Voices in the Shangshu,” Kai Vogelsang they betray the work of different authors or editors who held mutually opposed follows Kern’s lead in seeing two different political philosophies on display positions on questions of law and government” (p. 99). It is clear which of in the “Yao dian” (like Kern, he too notes the division of the chapter in the these two scenarios he favors by his characterization of it as being the one “Guwen” Shang shu into two separate chapters), but extends a similar analysis that “takes seriously” the difference, though one might imagine that other also to two other chapters: “Gaoyao mo” 皋陶謨 (Counsel of Gaoyao) and interpretations could also take it seriously. “Lü xing” 呂刑 (Punishments of Lü). Like the “Yao dian,” “Gaoyao mo” is In his concluding section, Vogelsang asks whether these differences he also divided into two separate chapters in the “Guwen” Shang shu: “Gaoyao detects in the three different chapters he analyzes might reflect a systematic mo” proper and “Yi Ji” 益稷 (Yi and Ji), of which “Gaoyao mo” purports to difference in the creation of the Shang shu, one arguing for “charismatic be a speech (or speeches) by Gaoyao 皋陶 , the minister of Shun and then Yu rule” and the other for bureaucratic government. However, combining the 禹 in charge of laws, and “Yi Ji” purports to be a speech (or speeches) by Yu, vocabulary and grammatical analyses that he brings to bear to the three who would succeed Shun as sovereign (and eventually to be recognized as the chapters, he finds “no such perfect division” (p. 102), and concludes “there 香港浸會大學饒宗頤國學院founder of the ). Juxtaposing the two parts of the chapter in parallel appear to be no consistent layers that recur in different chapters” (p. 103). columns A and B (based on the translation of ), Vogelsang From this he concludes “Given the complexity of the text’s history, this argues that despite their superficial similarity, they reveal different vocabulary picture is not at all surprising; in fact, any degree of uniformity would be a and grammatical usages, and, more important, that part A “centers on the virtue surprise. For modern scholars of the Shangshu, this means that they have to of the ruler” (p. 71) whereas part B seems to reverse part A’s arguments (p. analyze not only every chapter but also every paragraph by itself, taking into 72), though the reversal is hard to follow because of the fragmented nature account the larger contexts to which they may belong and also the numerous 422 423

interpolations, glosses that slipped into the text, and so forth. The manifold and compares the text to the recently discovered Tsinghua manuscript *Bao xun 保 perhaps competing voices in the Shangshu will continue to puzzle scholars” 訓 (Treasured instruction), for which he also provides a complete translation.5 (p. 103). This is certainly true, though I very much doubt that anyone who has Before then going on to translate the second half of “Gu ming,” Meyer taken the trouble to read the entirety of the Shang shu would be puzzled to find inserts a theoretical interlude, which he terms “Recontextualization and competing voices in it. Memory Production.” He says that by having a historical context “the text The third essay in the volume, by Dirk Meyer, examines another chapter transmutes the immediacy of the speech, and the communicative event that is divided into two chapters by the “Guwen” Shangshu: the “Gu ming” 顧 underlying the text is brought to the fore. When speech is recontextualized thus 命 (Testamentary charge) and “Kang Wang zhi gao” 康王之誥 (King Kang’s within a historical situation, real or imagined, it is no longer just reenacting the announcement). “Gu ming” purports to record the deathbed testament of King event—the speech now becomes a mediated, archived object. It is no longer Cheng of Zhou 周成王 and then the installation of his son 釗 as his performative or addresses the immediate witnesses of the occasion. Thus, successor, King Kang 康王 . However, unlike Kern and Vogelsang, Meyer the text that includes the speech becomes a reference tool, oral or written, proposes to follow the “Jinwen” 今文 (Modern text) Shangshu in reading the to inform a wider audience across the distance of time. It becomes a lieu two parts together as one consolidated whole, though, somewhat confusingly, mémoire, a place that stores—in fact, constructs—memory” (p. 127). It seems he refers to the two parts of the text as A and B and notes that they are very to me that this is just another way to say that the text is a text. Finally, Meyer different in nature. Discounting the purported historical context, he argues that closes with the following claim: “Gu ming” “is not primarily a document that records historical change through events but a multilayered intellectual enterprise where ancient communities I do not wish to say that “Gu ming” in its entirety was an Eastern Zhou

come to terms with the changing realities of their social and intellectual fabrication, as certain elements of the text may well be of a much older Book Reviews experience” (p. 106). In an understanding already familiar from his earlier date. But the making of the text as a lieu de mémoire to be used in publications, Meyer goes on to explain that he views the text as existing politico-philosophical discourse is clearly indicative of Eastern Zhou “primarily in oral form”: textual intervention (p. 138).

By constructing a narrative that can be understood as giving a reason (the This suggestion, which flies in the face of virtually all other datings of the emergency created by the king’s imminent death and, therefore, his need text,6 would be more persuasive if it were accompanied by specific evidence, to quickly name a successor) for its physical existence, it thus expresses especially linguistic or paleographic evidence. a sense of unease about, and therefore the need to explain, the event of The fourth essay, by Joachim Gentz, is entitled “One Heaven, One recording what may well (or should ideally?) exist primarily in oral form History, One People: Repositioning the Zhou in Royal Addresses to Subdued (p. 108).

After presenting an annotated translation of King Cheng’s address,4 Meyer then 5 In the case of this translation, the annotation is much lighter than in the case of Meyer’s “Gu ming” translation, and it is marked by a similar absence of reference to any relevant Western literature on the text, for which one might note Shirley Chan 陳慧 ,“Zhong 中 and Ideal 4 It is somewhat disconcerting that Meyer only rarely indicates any debt to previous English Rulership in the Baoxun 保訓 (Instructions for Preservation) Text of the Tsinghua Collection 香港浸會大學饒宗頤國學院translations: those by and Bernhard Karlgren (the glosses of whom are of Bamboo Slip Manuscripts,” Dao 11.2 (2012): 129–45, and Chapter Seven of ’s occasionally noted, especially in the linguistically more challenging second part of the Buried Ideas: Legends of Abdication and Ideal Government in Early Chinese Bamboo- chapter); the extensive translation given by Edward Shaughnessy (“The Role of Grand slip Manuscripts (Albany, NY: State University of New York Press, 2015), 263–314, which Protector Shi in the Consolidation of the Zhou Conquest,” Ars Orientalis 19 [1989]: 51–77) presents a complete annotated translation. seems to have avoided Meyer’s bibliographical search altogether, even though its central 6 Chinese tradition, of course, dates the “Gu ming” chapter to the , as does not argument that the “Gu ming” chapter underscores the consolidation of Zhou dynastic rule only almost all modern Chinese scholarship, but also most Western scholarship, including that anticipates Meyer’s main conclusion (as stated, for instance, on p. 137). of both Martin Kern and Kai Vogelsang (for which, see p. 110 n. 17 of Meyer’s essay). 424 425

Enemies in the ‘Duo shi’ 多士 and ‘Duo fang’ 多方 Chapters of the Shangshu the two “Duo” chapters are particularly well structured while the “Shang shi” and in the ‘Shang shi’ 商誓 Chapter of the Yi Zhoushu”; the title provides chapter is rather loose in its organization and rhetoric. a good indication of the contents of the essay. Gentz argues that these three He concludes the chapter with a discussion of the vexed question of the texts are unique in world literature insofar as they seek to persuade conquered date of the texts. Employing the sort of comparison of the Shang shu with foes by force of reason that the “historical rupture” of the conquest is actually defined as “continuity” (p. 174). He also suggests that the three chapters show a similar organization, which he outlines as follows (p. 157): “xian zhe wang” a standard phrase seen in such chapters of the Shang shu as “Kang gao” 康誥 , “Jiu gao” 酒誥 , and “Shao gao” 召誥 , but more importantly, shi 誓 is an attested allograph for zhe 哲 “wise” in Western Zhou bronze inscriptions, as shown by the parallel phrases in the • Indictments of recent transgressions of the subdued Shang following two inscriptions describing ancestors: • Historical precedent: the Xia were likewise overthrown by Shang • The infamous tyrant Zhou of Shang turned against Heaven’s order zhong: [...] was able to make wise his virtue. • Heaven thereupon transferred its Mandate to the Zhou 梁其鐘：克哲厥德 • Indictments of recent transgressions of the subdued Shang (in the form Pan Sheng gui: [...] was able to make wise his virtue. 番生簋：克誓厥德 of questions) • Order to cooperate combined with a threat to exterminate those who do Thus, Gentz’s translation is almost certainly correct; the graph 誓 in this text was but an early not obey allograph for 哲 . However, the reason that this point needs to be noted is, as noted already in the nineteenth century by Zhu Youzeng 朱右曾 (jinshi 1838), these three occurrences of the graph 誓 in the text of this chapter doubtless influenced the title of the chapter “Shang shi” 商

He then identifies thirty-one different “modules” that go into the making of Book Reviews 誓 , which Gentz (and also Kern) translates as “Harangue to Shang” (p. 155 and p. 286); for Shang shu chapters (pp. 160–61), many of which he finds to be unique to Zhu Youzeng’s suggestion, see Huang Huaixin, Zhang Maorong and Xudong, Yi Zhou the Shang shu, including both “Jinwen” and “Guwen” chapters. He notes shu huijiao jizhu, 477. On the same page as this translation, Gentz notes of the text: that the “Duo shi” and the “Duo fang” show a number of parallels with such other Shang shu chapters as the “Jun Shi” 君奭 (Lord Shi) and “Jiu gao” 酒 Although classified as a shi 誓 (harangue), it differs from the harangues in the Shangshu “ ” 誥 (Announcement on wine). Within the chapter he provides translations and in structure and diction but seems close to the two Duo chapters there (p. 155). analyses of the “Duo shi” and “Duo fang” chapters (and provides a complete This too is correct. However, he then goes on to treat all three of the texts as “harangues”: translation of the heretofore untranslated “Shang shi” chapter in an appendix, 7 In addressing this particular group, the “Shang shi” and “Duo” chapters therefore follow pp. 182–88 ), paying careful attention to rhetorical elements, suggesting that a rhetorical and compositional pattern also used in other Shangshu harangues when particular groups that are not in line with the king’s commands are admonished (p. 156). 7 Gentz’s translation of this difficult text is generally reliable. However, there is one point to note This compositional line is almost identical with the fixed sequence of elements that regarding the text that he used, which has important implications for the discussion of Shang Martin Kern describes in his analysis of the harangues (shi 誓 ) of the Shangshu (p. 158). shu chapter types in both Gentz’s own chapter and also that of Martin Kern,“ The‘ Harangues’ (Shi 誓 ) in the Shangshu” (for which, see below). As Gentz notes at the beginning of his translation, For his part, Kern says of the “Shang shi”: he uses the Yi Zhou shu text given in Huang Huaixin 黃懷信 , Zhang Maorong 張懋鎔 and This text presents a speech that King Wu purportedly gave to the captured Shang officers Tian Xudong 田旭東 , Yi Zhou shu huijiao jizhu 逸周書彙校集注 (: Shanghai Guji after the Zhou conquest; as such, it is not a battle speech but rather is similar to the “Many 香港浸會大學饒宗頤國學院chubanshe, 1995), 477–94. This edition, revised in 2007, is now generally accepted as the best Officers” (“Duo shi” 多士) and “Many Regions” (“Duo fang” 多方) chapters in the current critical edition, and so this was a reasonable choice. However, it should be noted that Shangshu (p. 286). Huang routinely “corrects” the received text whenever he perceives the need, only making mention of any original reading in his discussion. For instance, in three places where the This too is correct. Unfortunately, neither Gentz nor Kern has noted the accidental way that received text of the Yi Zhou shu reads “xian shi wang” 先誓王 , Huang’s text reads “xian zhe this chapter came to be termed a “shi” 誓 . The text is actually not a shi “harangue” at all, wang” 先哲王 , which Gentz translates as “former wise Shang king(s)” (twice on p. 183 and but rather a standard gao 誥 “announcement,” as are also both the “Duo shi” and “Duo fang” once on p. 186). There is considerable philological support for this emendation; not only is chapters. 426 427

Western Zhou bronze inscriptions previously used by Kai Vogelsang,8 and unnecessary to review them in detail here. However, there is one important finding a “lack of terminological and conceptual overlap” between them, Gentz substantive change in the two versions of Meyer’s essay, to which he himself suggests “bronze inscriptions do not help in dating the ‘Duo’ chapters” (p. calls attention,11 which should perhaps be highlighted. 177). Instead, he turns to what he terms “ideology” to suggest that the texts In the conclusion to his 2014 publication, Meyer argued that the form must have been written about the mid-Western Zhou period, some one hundred of the Zhou Wu wang you ji Zhou Gong suo zi yi dai wang zhi zhi manuscript years after the events of the conquest. Much of his argumentation in this regard suggests that it was intended to be circulated widely and read in private by seems to me to be counter-intuitive, as when he says on page 179 “The loyalty individuals. and goodwill of the vassals upon which the Shang and early Zhou rulers still depended are superseded by the goodwill of Heaven, which, however, is not Given its presentation of a dramatic narrative, the form of the “Zhōu reliable.” Indeed, the evidence of bronze inscriptions suggests that by the mid- Wǔwáng yǒu jí” seems suited to private text consumption. In this respect, Western Zhou, the descendants of Shang, to the extent that they can even still the opening frame of the text is particularly revealing. It presents a be identified in the inscriptional record, had been thoroughly assimilated into dramatic setting that brings to the fore all the contextual information the Zhou regime. This being the case, one wonders why the Zhou rulers would necessary for an indeterminate audience to confront the text and its have felt any need to try to persuade the Shang descendants to obey them. I message. The assumption that the text was not just produced for a known would not want to suggest that the received texts of the “Duo shi” and “Duo and limited group of recipients but that it was made available for wider fang” are among the earliest Zhou documents in the Shang shu, but it would be distribution and independent text circulation is further supported by the hard to imagine a better historical context for their initial formulation than the physical properties of the manuscript that suggest manuscript production

immediate aftermath of the Zhou conquest itself. on a larger scale and not just for this one instantiation of the text. Book Reviews The fifth and sixth chapters, by Magnus Ribbing Gren and Dirk Meyer Despite my conviction that the “Zhōu Wǔwáng yǒu jí” was read respectively, both address the “Jin teng” 金縢 (Metal-bound coffer) chapter of in private by individuals, which has significant implications for our the Shang shu and particularly its counterpart among the Tsinghua manuscripts, understanding of reading and knowledge transmission in the late Warring there entitled Zhou Wu wang you ji Zhou Gong suo zi yi dai wang zhi zhi 周 States period, I maintain that the text also had a politico-philosophical 武王有疾周公所自以代王之志 (The record of King Wu being ill and the dimension that goes beyond plain Zhōu propaganda to portray the Duke Duke of Zhou substituting himself for the king). Since both of these essays are of Zhōu as a loyal statesman in selfless service to his lord.12 essentially republications of studies that were previously published in widely available journals,9 in Ribbing Gren’s case “with minor modifications,” and On the other hand, the essay in the book under review, published in 2017, in the case of Meyer’s essay with numerous stylistic changes,10 it seems to me presents a completely different context for the production and reception of the text in ancient China. 8 Kai Vogelsang, “Inscriptions and Proclamations: On the Authenticity of the ‘Gao’ Chapters in the ,” Bulletin of the Museum of Far Eastern Antiquities 74 (2002): 138–209. [...] “Zhou Wu Wang you ji” is more than a narrative: it is performative in 9 Magnus Ribbing Gren, “The Qinghua ‘Jinteng’ 金縢 Manuscript: What It Does Not Tell Us the sense that it requires its audience to activate the historical background about the Duke of Zhou,” T’oung Pao 102.4–5 (2016): 291–320; Dirk Meyer, “The Art of stored in cultural memory. In its elegant symmetrical brevity, which also Narrative and the Rhetoric of Persuasion in the ‘*Jīn Téng’ (Metal Bound Casket) from the constitutes its fundamental incompleteness, the text stages itself as an 香港浸會大學饒宗頤國學院Tsinghua Collection of Manuscripts,” Asiatische Studien / Études Asiatiques 68.4 (2014): aesthetic artifact that becomes fully meaningful only in its reception by a 937–68. community of insiders. 10 Aside from ubiquitous rephrasings of essentially the same formulations, the essay in the book This, to me, is indicative of another feature. The fact that “Zhou under review here, “‘Shu’ Traditions and Text Recomposition: A Reevaluation of ‘Jinteng’ 金 縢 and ‘Zhou Wu Wang you ji’ 周武王有疾 ,” includes one additional section, entitled “Text and Fabula,” and eliminates from the earlier essay several of the sorts of diagrams that Meyer 11 See n. 12 below. has used in many of his previous publications. 12 Meyer, “The Art of Narrative and the Rhetoric of Persuasion in the ‘*Jīn Téng,’” 962–63. 428 429

Wu Wang you ji” speaks to the meaning community of beholders of employ a “formalistic approach” (p. 267), differentiating among “dramatic” hegemonic Zhou culture and memory insinuates that successful text speeches (nineteen of the twenty-eight chapters of the Shang shu but only three reception requires some degree of acquiescence with regard to the chapters of the Yi Zhou shu), which “are emotionally laden and personalized conclusion in “Zhou Wu Wang you ji,” and thus to the values propagated and have a richer repertoire of emphatic devices”; “nondramatic” speeches by the Zhou. It thus appears that a community with severe skepticism (only one chapter of the Shang shu, i.e. the “Hong fan” 洪範 [Vast plan], but about the motives of the Duke of Zhou cannot constitute the target sixteen chapters of the Yi Zhou shu), which “appear as treatises superficially audience of “Zhou Wu Wang you ji.” More likely, “Zhou Wu Wang you furnished by emphatic devices reminiscent of dramatic speeches” (p. 255); ji” was “preaching to the converted,” that is, a community that subscribed “brief speeches related to dream revelations” (three chapters of the Yi Zhou to orthodox Zhou values while also being aware of, and perhaps even shu)(p. 270); “texts with writing-informed contextualization” (one chapter of sharing, some of the doubt over the duke’s position in Zhou cultural the Shang shu [“Lü xing”] and three chapters of the Yi Zhou shu), texts that memory. [...] explicitly mention their own composition (p. 270); “plot-based stories with The actuality of “Zhou Wu Wang you ji”—a text that embodies the dialogues” (only the “Jin teng” of the Shang shu, as well as two chapters of experience of doubt and belief on the part of audiences—thus clearly lies the Yi Zhou shu)(p. 272); and “texts with speeches that are difficult to classify” in performance, that is, the formalized reenactment of that incident of (five chapters of the Shang shu and six chapters of the Yi Zhou shu)(p. 273). disbelief. The structural and visual properties of the text seem well suited That this sounds rather like the Borgesian Celestial Emporium of Benevolent to oral performance,13 possibly including the marking up of breathing Knowledge seems not to be lost on Grebnev. In a note expressing his gratitude points in the manuscript, which would have facilitated intoning the text to Robert Eno, one of the press referees for the book, Grebnev makes the

aloud (pp. 244–46). following allowance: Book Reviews

That virtually the same analysis of the text could, within the span of three years, My classification may appear overly mechanical since I prioritize the easily produce such diametrically different conclusions about its social context might identifiable formal textual features over the more intricate ones and largely cause one to question just how firmly grounded these conclusions might be. disregard the contents of texts. Indeed, it would be possible to make my Chapter Seven, “The Yi Zhoushu and the Shangshu: The Case of Texts study more fine-grained by highlighting the important differences in texts with Speeches,” by Yegor Grebnev, attempts to identify analogous groups that otherwise appear formally similar. This improvement, however, would of texts within the Yi Zhou shu and the Shang shu, two collections that have have required me to rely more on subjective interpretation and would have traditionally been understood as containing texts of the same genre or genres.14 made my classification lose much of its transparency, which I prefer to Building on his recent Ph.D. dissertation,15 he focuses on texts that purport to avoid. Nonetheless, I acknowledge that a scrutiny of textual intricacies be speeches or predominantly to feature speeches, which as he notes include all would complement my detached formalistic analysis and perhaps challenge but one the chapter of the “Jinwen” Shang shu (the one chapter that he regards, it in important ways (p. 253 n. 13). understandably, as essentially narrative is the “Yu gong” 禹貢 [Tribute of Yu]) and about thirty-four of the fifty-nine extant chapters of the Yi Zhou shu. Since In a concluding section entitled “Discussion,” he notes: this encompasses almost the entirety of the two texts, he proposes further to All this makes the Yi Zhoushu appear very different from the Shangshu, 香港浸會大學饒宗頤國學院 which contains predominantly dramatic speech. It therefore seems that in 13 The original here contains a note (n. 61): “Here I correct my earlier remarks in Meyer 2014a.” the Shangshu an attempt was made to ensure the typological consistency 14 Instead of “genres,” Grebnev refers to “textual types,” adding a note (p. 251 n. 9) arguing of the collection. At the same time, the Yi Zhoushu appears more like an that his term is “perhaps more suitable for fine-grained analysis of textual structures in their diachronic evolution.” I fail to see the rationale behind this distinction. assemblage of very different texts that, if one follows the logic of its last 15 Yegor Grebnev, “The Core Chapters of the Yi Zhou shu” (Ph.D. diss., University of Oxford chapter, “Zhoushu ” 周書序 , had only one thing in common: they were 2016). believed to be somehow related to the (pp. 276–77). 430 431

This characterization is certainly unobjectionable, but it is also common separate such repertoires, that at some point during the late Warring States knowledge. One might hope that Grebnev will build upon his doctoral gave rise to the formation of the texts in a form recognizably related to dissertation to offer a more subjective “scrutiny of textual intricacies”; even at their received versions (p. 306). the risk of sacrificing transparency, this would doubtless show more about the diversity of the Yi Zhou shu. This argument is also unobjectionable, but it is hardly novel. Already on page Chapter Eight, “The ‘Harangues’ (Shi 誓 ) in the Shangshu,” is a 288, Kern notes that virtually all modern scholars have dated the texts to the second contribution by Martin Kern. As the title suggests, he examines the Eastern Zhou, and again at page 305, with particular respect to the “Mu shi,” six battlefield speeches or “harangues” (shi 誓 ), one of the five genres into which is the focus of the second half of his study, Kern notes that Herrlee G. which the Shang shu chapters have traditionally been divided. Kern focuses Creel declared the “Harangue at Mu” “almost certainly not what it purports in particular on three of these: the “Gan shi” 甘誓 (Harangue at Gan), which to be, a speech by King Wu,” and continues by saying “The same conclusion purports to be a speech by Yu at the beginning of the Xia dynasty; the “Tang was shared by , Zhang Xitang, and Matsumoto Masaaki, who all shi” 湯誓 (Harangue of Tang), attributed to Tang 湯 , the founder of the Shang suspected the text to date from the Warring States.” dynasty, before his troops attacked Jie 桀 , the last of the Xia kings; and the “Mu A second argument made by Kern is that these texts were performed as shi” 牧誓 (Harangue at Mu), said to have been the speech of part of later Zhou ritual celebrations. 周武王 on the morning of the battle at Muye 牧野 , in the course of which the Zhou overthrew the Shang. Kern provides complete translations for each of It is very possible that this colorful iconography of King Wu wielding these three short chapters. his yellow ax and brandishing his white banner had entered the historical

Beginning with the observation by Thucydides that his method in imagination via an early performance tradition. This tradition may have Book Reviews recording the battlefield speeches included in his The Peloponnesian War is “to developed in the choreography of the Zhou ancestral sacrifices [...] (p. make each speaker to say broadly what I supposed would have been needed on 300). any given occasion,” 16 Kern argues throughout the chapter that the battlefield speeches or harangues in the Shang shu were also made up retrospectively. In the Zhou ancestral temple, every such reperformance of King Wu’s speech evoked the imagined original harangue as the dynasty’s primordial I consider them not to be the actual words spoken but those that were call into being [...] (p. 308). retrospectively imagined (p. 284). Whatever the case, from the fact that Warring States and Han authors [A]ll of these should be considered not as documentary accounts but as knew about both a “Great Harangue” and a “Harangue at Mu,” it appears idealizing constructions of “remembered history” (p. 285). that more than a single version of King Wu’s battle speech existed, and that these different versions gradually found their way into different It frees us from the traditional—and methodologically unsustainable— written accounts. It is also possible that to the very end of the Zhou assumption that the received texts are discrete entities that in some state, until 256 BCE, the conquest of 1046 BCE was still remembered in inexplicable way survived over centuries in more or less pristine form performance (p. 311). and that therefore can be dated individually on the basis of their linguistic 香港浸會大學饒宗頤國學院 properties (p. 304). This is certainly possible, but as far as I know there is no evidence, whether textual or artifactual, showing that any of these harangues actually figured in [A]ll these texts were part of a common repertoire, or possibly of several any Zhou ritual. Indeed, Kern himself notes regarding the “Mu shi,” which once again is the main topic of his discussion throughout the second half of 16 Thucydides, The Peloponnesian War, trans. Martin Hammond (Oxford: Oxford University his study, “its early reception does not appear to have begun before the Han Press, 2009), 12 (1.22), quoted on p. 281. dynasty” (p. 288). Thus, while rituals certainly took place and performances 432 433

may well have been a part of them, one can only imagine a role for the discourse.” Elsewhere, Schaberg, like Kern, engages the imagination, harangues in them. imagining some sort of a time machine tape recorder that will allow us to Chapter Nine, by David Schaberg, “Speaking of Documents: Shu Citations recapture the sound of antiquity: in Warring States Texts,” is the odd man out among the chapters of this book in that it does not directly address any chapter or chapters of the Shang shu, Just as it may help us to imagine a complete anthology of archaic and but rather examines echoes (the sonic metaphor very much intended) of them archaizing passages, it may help to imagine the possibility—still just in Warring States and Qin and Han texts. Since Schaberg revisits topics that he barely out of our reach, but perhaps not for long, given the progress being discussed a decade and even two decades ago it would be pointless here—in a made in historical phonology and in the digital marking of texts—of review of studies of the Shang shu per se—to devote much space to discussing comparing any given string of sounds (the sounds designated by a series his broader concerns, which to a very great extent are echoes of his own past of graphs in a Shu citation, for instance) with a large number of other scholarship. strings within a defined corpus such as the proposed archaistic anthology. As in his celebrated A Patterned Past: Form and Thought in Early The aim of collation would be, not to bypass the written language, ,17 Schaberg writes so beautifully that it is easy to be but to identify those places where different choices in writing obscure enticed by his rhetoric. But consider the gulf between the first two sentences of fundamental identity in the words and sentences themselves. Archaism his essay and the last two: was not, at least in the period under consideration, a phenomenon that took place entirely in writing, without reference to the spoken word. Only Considered—naively—as a record of linguistic development, the collation of the type described here can make sense of Warring States

canonical early Chinese texts imply a strikingly discontinuous sort of period data, which make it clear that writers knew of some archaic writings, Book Reviews evolution. Between texts associated with the Western Zhou, such as the including some of the materials they called Shu, primarily through sound older portions of the Shangshu 尚書 and of the Shijing 詩經 , and Warring rather than through sight” (pp. 322–23). States texts like the 左傳 , there lies a dark age of change, in which syntax and vocabulary were transformed (p. 320). This may be interesting to imagine, but it is after all, still imaginary. The next two chapters return us to the Shang shu proper. Both Chapter Ten The Shu had not simply become the Shangshu: one can assume that it and Chapter Eleven, by Yuri Pines and Michael Hunter respectively, address the had, but doing so creates problems rather like those facing Ptolemaic “Wu yi” 無逸 (Without ease) chapter. I will consider the two studies separately, systems of astronomy; one can make the equations work, to be sure, though it is curious that whereas Pines notes at the outset of his study that “Wu but there is another, simpler way. If this alternative way of proceeding yi” is usually regarded as one of the less noteworthy of the chapters that purport calls into question the status of writing in pre-imperial China, it has the to date to the early Western Zhou and specifically to report the words of the compensatory advantage of evoking a world of living speech in which the Duke of Zhou, Hunter instead says “both substantively and formally ‘Wu yi’ is sound of antiquity (a sound often enough referred to as if it were silent, an among the most philosophical contributions to the ‘Shu’ tradition” (p. 394). unpronounced Shu) was still a part of intellectual discourse (p. 356). Pines indicates that his own interest was drawn to the text because of its use of historical precedent as a model for current governmental activity. A writer who can write so well must make sense, but I would suggest that no He begins by providing a complete annotated translation of the text. The text 香港浸會大學饒宗頤國學院equation that is currently known, no matter how complicated or how simple, points to four past sovereigns as paragons of good rule: Kings Zhongzong is sufficient to bridge the gap between this “dark age of change” and “a world 中宗 , Gaozong 高宗 , and Zujia 祖甲 of the preceding , and of living speech in which the sound of antiquity was still a part of intellectual 周文王 , all four of whom were characterized by their care for the common people. Pines stresses that the latter three of these are said 17 David Schaberg, A Patterned Past: Form and Thought in Early Chinese Historiography to have lived among the common people before coming to power, and that (Cambridge, MA: Harvard University Asia Center, 2001). this experience influenced their rule. Although the notion of protecting the 434 435

people (bao min 保民 ) is central to most Shang shu texts, Pines argues that begins by comparing “Wu yi” (and indeed many other chapters of the Shang the “Wu yi” is unique—at least in relation to the nine chapters that are usually shu as well) to wisdom literature from throughout the ancient world, especially regarded as dating to the Western Zhou18—in that it does not mention any role wisdom literature that takes the form of advice from an older man to a younger for ministers or advisers as coming between the king and the people. Another man. Other than the probably obvious point that old men have been giving feature of the text makes it unique within the context of Warring States political their juniors advice for a long time, I am not sure that the almost ten pages of philosophy: its emphasis on the place of manual labor in the education of the juxtaposed sayings shed much new light on any of these texts. proper sovereign. According to Pines, most Warring States thinkers encouraged When he turns to “Wu yi” itself, Hunter presents a subtly different rulers to enjoy their ease, so as to allow the bureaucracy to function without understanding from that of Pines. Whereas, as we have seen, Pines argues interference from them. From this, he concludes this section of his essay with for the virtue of rulers engaging in manual labor early in their lives, Hunter the statement that “contrary to our expectations of a canonical scripture, ‘Wu suggests that the text urges instead only that rulers “understand” (zhi 知 ) the yi’ appears not as representative of the mainstream ideological orientation of labors of others. Both Pines and Hunter explicitly note their disagreement the Eastern Zhou period but rather as a marginal—and in some respects even with each other, Pines at page 368 note 29 and Hunter at page 406 note 42; subversive—text” (p. 381). however, despite the great importance that the distinction between knowledge Pines concludes his study with what he labels frankly as “conjectures” (zhi) and action (xing 行 ) would take on in later Confucian philosophy, concerning the date of the “Wu yi” chapter. Failing to find any tell-tale including in exegesis of the Shang shu, in this case the difference between the linguistic features marking it as either of Western Zhou or Warring States two interpretations strikes me as superficial rather than crucial. pedigree, he suggests that the years immediately following the fall of the To me, a more important distinction might be termed historiographical

Western Zhou in 771 B.C., when the royal court was in the process of moving (though also with obvious political implications). Whereas Pines suggests Book Reviews east but had not yet consolidated itself might have provided the sort of that the sovereigns experienced at labor could serve as examples for their historical context that would motivate such a text. In this regard, he notes that successors, Hunter argues that “idleness is not simply the opposite of hard the recently published Tsinghua manuscript *Xinian 繫年 (Annals), of which work but also its indirect result” (p. 405). The “Wu yi” describes active rulers he has published one of the Western-language studies,19 might explain why for early in the dynasty, each of whom enjoyed lengthy reigns; on the other hand, the “tiny polity” that the Zhou king ruled at the time “the ‘lowly people’ could the more recent rulers toward the end of the dynasty “were idle during their have been more significant than focusing on officials and on regional lords, lives [...] some reigned for ten years, some for seven or eight years, some for who ‘ceased attending the Zhou court’” (p. 386). I am sure that Pines would five or six years, some for four or three years.” This reads like a precursor to readily admit to the speculative nature of this conclusion, but as he says in his the “dynastic cycle” argument that sees dynasties founded by vigorous and “Epilogue,” “this short essay suffices to show that a fresh look at the millennia- virtuous individuals, only to end with weak and willful rulers; just as sons of old documents may bring about new understandings” (p. 388). successful families have a tendency to be indolent, squandering the families’ Michael Hunter places “Wu yi” within a much broader context than does wealth, so too do successors of vigorous rulers tend to be idle, eventually Pines. Perhaps motivated by the editors’ quotation in the Introduction of Max bringing the dynasty to an end. Hunter concludes his discussion of this point Müller’s saying “He who knows one [religion], knows none” (p. 10), Hunter by terming “Wu yi” “a metadiscursive comment”:

In short, I propose that we read “Wu yi” as a metadiscursive comment 誥 大誥 康 香港浸會大學饒宗頤國學院18 Pines regards the “five gao ” (i.e., “Da gao” [Great announcement], “Kang gao” on one of the core values of the instructions genre. On the one hand, it 誥 [Announcement to Kang], “Jiu gao” 酒誥 [Announcement on wine], Shao gao 召誥 carves out a rhetorical space for kings and noblemen to justify the pursuit [Announcement of Shao] and “Luo gao” 洛誥 [Announcement at Luo]), as well as the “Duo of pleasure. On the other, “Wu yi” highlights a tension within the generic fang” 多方 (Many regions), “Duo shi” 多士 (Many sires), “Zi cai” 梓材 (Catalpa timber) and “Jun Shi” 君奭 (Lord Shi) as representative of Western Zhou ideology. didacticism of the Shangshu and related texts: the very virtues that are 19 Yuri Pines, “Zhou History and Historiography: Introducing the Bamboo Manuscript Xinian,” conducive to the material success of one generation indirectly contribute T’oung Pao 100.4–5 (2014): 287–324. to the vices of the next (p. 407). 436 437

In his conclusion, Hunter uses this characterization to offer the following On the other hand, the text’s place in the two different sequences of the Shang suggestion, almost in passing, concerning the date of the text: shu would suggest a later date. In the “Jinwen” Shang shu, it comes between the “Gu ming” and “Lü xing” chapters, which is to say sometime between the [...] the metadiscursive qualities of “Wu yi,” a philosophical text in “Shu” succession of King Kang and the reign of King Mu 穆王 , while in the “Guwen” clothing, point to a much later date of composition, perhaps in the late Shang shu, the text comes in the penultimate position, just before the “Qin shi” (p. 413). 秦誓 (Harangue of Qin), the contents of which date it to the reign of 秦穆公 (r. 659–621 B.C.). There is nothing in the content of the “Bi If, as Pines suggests (and as noted above), the “Wu yi” betrays no tell-tale shi” that allows this dating to be resolved: the speaker is identified only as a linguistic signs of a Warring States composition, one wonders if the form of “duke” (gong 公 ), the text later implying that he is a duke of Lu, since it is the text alone is sufficient to support such a broad-reaching conclusion. addressed to “men of Lu” (Lu 魯人 ). Attacks on the Huai Yi and Xu Rong With Chapter Twelve, Maria Khayutina’s “‘Bi shi’ 粊誓 , Western Zhou are no more helpful in pinpointing a date, since they occurred intermittently Oath Texts, and the Legal Culture of Early China,” the scholarship takes throughout the entire Western Zhou dynasty and also into the Spring and a distinct turn from metadiscursive comment to local history. Khayutina Autumn period. Khayutina surveys the evidence for these wars, but noting that begins with a translation of this short text of only 180 characters, which she “Wars between Lu and the Yi of Huai during the later admits poses no linguistic difficulty. She begins her discussion with the are not attested,” she concludes rather simplistically that “the Spring and problem of the title. She refers to the text as “Bi shi” 粊誓 , said to have been Autumn period as a whole does not qualify as the historical setting for ‘Bi shi,’ the title in the “Guwen” version of the text, even though in almost all received which supports the traditional view of ‘Bi shi’ as depicting the war under King 費誓 費 editions the title is written as (in which case the , normally read as fèi, Cheng” (p. 424). When she turns to an examination of the language of the text, Book Reviews is also to be read as bi, normally understood to refer to a place name found in she offers a very different date. present-day Shandong 山東 province), which is the reading of the “Jinwen” In this examination of the language, Khayutina first compares it to oaths Shang shu. As for the shi 誓 of the title, which elsewhere in the Shang shu recorded in three different Western Zhou bronze inscriptions, all of which is used for battlefield speeches or harangues (as studied by Martin Kern in date to the late Western Zhou. While the language of the “Bi shi” is similar Chapter Eight of this volume), she argues on the basis of Western Zhou bronze to that in these bronze inscriptions, it also differs in two important respects. inscriptional usage that it should be understood instead as an “oath,” noting First, it enjoins the recipients to “listen” (ting 聽 ) to the oath, which is never that unlike other shi of the Shang shu, in which speakers state that they are seen in the inscriptions but which is also mentioned in the “Qin shi,” as well making a shi, in this case the character appears only in the title. Although it is as being regularly mentioned with respect to oaths in such later texts as the true that much of the text reads like oaths in those bronze inscriptions (as well Zuo zhuan 左傳 . Second, it refers to administrative districts jiao 郊 (suburbs) as in later texts), it is perhaps understandable that later editors of the Shang shu and sui 遂 (marches), which are never seen in Western Zhou inscriptions and would have regarded it as a battlefield speech, since the text begins by calling which seem to be distinctly anachronistic for the Western Zhou. She then goes troops to march against the Huai Yi 淮夷 and the Xu Rong 徐戎 , and then on to compare the chapter to the poem “Bi gong” 閟宮 (Secret palace) in the returns at the end to mention the exact date of the campaign. Nevertheless, the Lu song 魯頌 (Lu liturgies) section of the Shi jing 詩經 (Classic of poetry), question of this title provides the context for Khayutina to provide a detailed probably the latest surely datable poem in the Shi jing (it is at least implicitly discussion of oaths and their role in the early development of penal law in dated to the time of Duke Xi of Lu 魯僖公 [r. 659–627 B.C.]), parallels with 香港浸會大學饒宗頤國學院China. which have led many modern scholars to date the “Bi shi” as well to the She next notes that there have been at least two and perhaps three seventh century B.C. Finally, Khayutina notes the Zuo zhuan recounts that different understandings of the text’s historical context: the “Preface to the the state of Lu first formulated a legal code, or at least a public rationale for Documents” (Shu xu 書序 ) attributes it to Bo Qin 伯禽 , the eldest son of when punishments could be used, during the reign of Duke Xi’s grandson, the Duke of Zhou and the putative founder of the state of Lu 魯 , and more Duke Xuan of Lu 魯宣公 (r. 608–591 B.C.), and that there are also parallels specifically to the civil war at the very beginning of the Western Zhou dynasty. between the language of “Bi shi” and these legal codes. All of this leads her to 438 439

conclude, quite reasonably, it seems to me, that the “Bi shi” must have been it is not surprising that later editors and scholars should have sought to give it written about this time, just as its placement in the “Guwen” sequence of the an even more ancient pedigree. It seems to me that if cultural memory explains Shang shu would suggest. anything, it would be this later editorial urge. However, not content just to conclude when the “Bi shi” might have been The penultimate chapter in the volume, “Concepts of Law in the written, she goes on to speculate as to why and even how it may have been Shangshu,” is by Charles Sanft. He starts with the observation that notions of written. Without ever explicitly mentioning the work of Jan Assmann (one of law and legal practice exhibit both consistency and difference throughout the whose books is listed in her bibliography20), she frequently resorts to the notion several chapters of the text that are usually referred to in discussions of the of “cultural memory” for which he is well known, and says that the author or law. The two most important of these chapters are surely the “Kang gao” 康 authors of “Bi shi” chose “the ancient form of the ‘oath’ and by introducing 誥 (Announcement to Kang) and “Lü xing,” the latter of which is Sanft’s main some expressions resembling the language of Western Zhou ‘commands,’ they focus (though he considers concepts of law much more broadly throughout the possibly attempted to produce a text that their contemporary readers would entire text of the Shang shu). He suggests that whereas “Kang gao” presents recognize as an ancient document” (p. 430). She even goes so far as to give a an idealized view of law as dependent on virtue and the selection of the proper name to the author—or at least to the circle of authors: Zang Wenzhong 臧文 person to decide legal cases, “Lü xing” posits a system of jurisprudence and 仲 (d. 617 B.C.), famed from his appearances in the Zuo zhuan and Guo yu 國 thus is “the Shangshu chapter that best reflects attitudes toward legal practice 語 , and also as a (negative) paragon for . in China throughout its history” (p. 471), even though such an important traditional commentator as Cai 蔡沈 (1167–1230) dismissed the text as a Technically, producing texts similar to “Bi shi” was not a difficult task for warning of how not to rule (p. 464).

learned men of Zang Wen zhong’s circle. As a minister, he certainly had Sanft explores several different topics in which “Lü xing” may be viewed Book Reviews access both to the archives of Lu and to the ducal temple where bronze as exceptional within the context of the Shang shu: “the role of legal practice vessels with inscriptions commissioned by former Dukes of Lu were within society” (i.e., whether law and especially punishment are necessary); preserved. If these archives or vessels contained some records from the types of punishment and especially the role of monetary redemption in first reigns of Lu, they would be similar to early Western Zhou bronze mitigating those punishments; the place of doubt in legal processes; and inscriptions from other places. Thus, they would differ from the current intent. With respect to some of these topics, Sanft argues, against traditional text of “Bi shi” in the ways discussed above. Quite certainly, the archives commentators, that “Lü xing” is inconsistent with other Shang shu chapters, also contained many more recent command and oath texts, similar to and in other cases, again against traditional commentators, that it is in fact those in middle to late Western Zhou bronze inscriptions. After studying consistent with the other chapters. Although I am not convinced by all of his these materials, “ancient documents” could easily be fabricated (p. 438). analysis, it is refreshing to note that the discussion is thoroughly grounded in both the texts and traditional commentaries. I will consider just two of these It seems to me that there is little doubt that the “Bi shi” is a document of the topics below, one case in which I am unpersuaded by Sanft’s reading and another early Spring and Autumn period, just as Khayutina concludes. However, I in which I think he corrects an important misinterpretation of “Lü xing.” see no archaicism—conscious or otherwise—and certainly no “fabrication” One of the most common early Chinese assumptions about law in general in either the language or the ideas of the text. Rather, the author was simply and punishment in particular is that while they are necessary, to be sure, still writing in the language of his own day, which, after all, did not yet differ very they are necessary evils. Sanft suggests that “Lü xing” may be an exception to 香港浸會大學饒宗頤國學院dramatically from that of the late Western Zhou. Why this occasional text this critical attitude toward punishment. He regards as particularly important in of little apparent historical or philosophical significance should have been this regard a couple of uses in the “Lü xing” of the apparent oxymoron “xiang included in the Shang shu is something of a mystery, but once it was included xing” 祥刑 , which he translates as “beneficial punishment.” He considers, but dismisses, Zheng Xuan’s 鄭玄 suggestion that xiang 祥 (auspicious) be read 20 Jan Assmann, Cultural Memory and Early Civilization: Writing, Remembrance, and Political as xiang 詳 (to examine carefully), arguing that this reading is too easy and Imagination (Cambridge and New York: Cambridge University Press, 2011). renders the sense of the text mundane. Perhaps it is mundane, but it seems to 440 441

me to be the reading also of the canonical Pseudo- commentary as literary tradition that positively revel in the abundance of China’s geographic well: gao ru yi shan yong xing zhi dao 告汝以善用刑之道 , which Sanft cites diversity.21 as support for his reading. Sanft’s translation of this sentence as “I will tell you The “Yu gong” of course has had a central place in modern Chinese of the way of using punishments well” (p. 455) is certainly unobjectionable, study of historical geography, inspiring a journal of the same title, published but I would argue that “using punishments well” is something very different between the years 1934 and 1937 and edited by 顧頡剛 (1893– from “beneficial punishment.” 1980) and Tan Qixiang 譚其驤 (1911–1992), the fathers of modern Chinese On the other hand, I think Sanft’s discussion of another topic introduces an historiography and historical-geography respectively. McNeal does not add important corrective to the views of traditional commentators. As noted above, much to standard introductions to the text, even in recent Western languages.22 Cai Shen criticized “Lü xing” for explicitly allowing corporal punishments to His account of the “Wang hui” chapter of the Yi Zhou shu, on the other hand, be reduced against the payment of monetary fines. Commentators throughout does constitute a real contribution to our understanding of early Chinese history have echoed Cai’s criticism of this apparent monetization of the legal literature. This text, Chapter 59 in the Yi Zhou shu, is fascinating for all process. However, Sanft argues convincingly that “Lü xing” makes such sorts of reasons, not least of which is its description of the flora and fauna allowance only in cases of doubt, and in this way is consistent with passages purportedly brought to the Zhou court early in the reign of King Cheng of elsewhere in the Shang shu (such as in the “Da Yu mo” 大禹謨 chapter) that Zhou. Unfortunately, aside from a pair of strange canines (one of which has a have long been praised as the epitome of fairness. This does not mean that “Lü face like a man’s and is able to talk, and the other that is able to fly and to eat xing” is invariably consistent with the other chapters (as Sanft notes, it seems tigers and leopards), none of these is presented here. McNeal’s purpose is with to take no notice of intent in the adjudication of cases), but it is surely not bigger issues, as when he concludes:

exceptional within the Shang shu. Book Reviews

The final chapter of the book, Robin McNeal’s “Spatial Models of the 21 I am delighted to see this English-language introduction to the “Wang hui” chapter, one of the State in Early Chinese Texts: Tribute Networks and the Articulation of Power most interesting chapters in the Yi Zhou shu, and one that has loomed large in the scholarship and Authority in Shangshu ‘Yu gong’ 禹貢 and Yi Zhoushu ‘Wang hui’ 王會 ,” on that text. However, it is an unusual choice to compare with the “Yu gong” chapter of the takes us back—if not quite to metadiscursive comment—then certainly to big- Shang shu. The “Wang hui” chapter is usually compared with the Shan hai jing 山海經 (Classic picture questions. He begins with three such questions: of mountains and seas; translated by McNeal as Itineraries through Mountain Ranges and Waterways; p. 483) for its descriptions of strange flora and especially fauna from foreign lands. “ ” Recent scholarship has highlighted many of the fundamental uncertainties The chapter of the Yi Zhou shu usually compared with the Yu gong chapter, on the other hand, is the “Zhi fang” 職方 “Officering the regions; Ch. 62), a text unmentioned by McNeal. scholars face when trying to describe the early Chinese state in historical The “Zhi fang” chapter shares much of its content and language in turn with the “Zhifang shi” terms. What was the scope and structure of early Chinese states? To what 職方氏 (Officers of the regions) entry of the “Xia guan” 夏官 section of the Zhou li 周禮 ; the extent can models useful in describing state formation in other parts of the question of the priority of these two texts is too complicated to treat here, but for a vigorous world be applied fruitfully in China? And just how many states were there defense of the Zhou li’s priority, see Huang Peirong 黃沛榮,“ Zhou shu yanjiu” 周書研究 (Ph.D. in the formative period? (p. 475) diss., National University, 1976), 244–64. Like the “Yu gong” chapter, the “Zhi fang” chapter also describes a nine-fold division of the Chinese world, the names of the regions He begins his discussion by characterizing traditional historiography as being largely identical (the “Yu gong”’s Xu 徐 and Liang 梁 are replaced with You 幽 and portraying a monolithic civilization, against which he pits the archaeological Bing 並 in the “Zhi fang” chapter). What is more, the similarity between the “Yu gong” and 香港浸會大學饒宗頤國學院record, which he suggests “gives us a strikingly different picture,” one marked the “Zhi fang” chapters extends well beyond the spatial orientation and includes also similar descriptions of local products. primarily by diversity, especially (but by no means only) geographic in nature. 22 See the accounts in Mark Edward Lewis, The Flood Myths of Early China (Albany, NY: State It is geography that is McNeal’s primary interest in this chapter, though it is University of New York Press, 2006), Chapter Five; and Vera Dorofeeva-Lichtmann, “Ritual somewhat jarring to note that the two texts on which he focuses, the “Yu gong” Practices for Constructing Terrestrial Space (Warring States–Early Han),” in Early Chinese 禹貢 (Tribute of Yu) chapter of the Shang shu and the “Wang hui” 王會 (Royal Religion, pt. 1, Shang through Han (1250 BC–220 AD), eds. John Lagerwey and Marc convocation) chapter of the Yi Zhou shu are the two texts in the early Chinese Kalinowski (Leiden: Brill, 2009), 595–644. 442 443

This vision of imperial power, imagined and projected onto a legendary Shangshu chapters, explores the rhetorical patterns of cultural memory, past, became a model for what we would now call foreign relations, and examines specific political ideas against a multiplicity of possible a clear example of how the historical imaginary came to create and historical contexts, from the founding days of the Western Zhou through, maintain a political reality. One of the best measures of the importance nearly a thousand years later, the early empire. Our conclusions are often the Shangshu has had over the last two millennia is precisely its ability to unexpected to the point of overturning the accumulated wisdom of two enforce on the future its vision of the past (p. 491). millennia [...] (pp. 8–9).

*** Not one of these claims can withstand scrutiny. The first statement regarding the “forged” Shangshu is curious; the matter is disposed of in the Introduction in one sentence, and nothing else in the book engages with the post-Han At the beginning of this review, I noted that the book is marred only by transmission of the Shang shu at all, and certainly not with the debate over the the Introduction, which I characterized as marked by a “self-congratulatory “Ancient Text” (guwen 古文 ) versus the “Modern text” (jinwen 今文 ) versions triumphalism,” both denigrating past scholarship and making claims about the of the text, for which one would be far better served by re-reading Paul scholarship in this book that are well beyond what the book delivers. Consider Pelliot’s 1916 article “Le Chou King en caractères anciens et le Chang Chou che just the following three statements. wen.”23 The second quotation with its accusation that past scholars have used “a text one has not truly studied” can only be said to be shocking. Leaving aside The question of the “forged” Shangshu in ancient script deserves further the great many Chinese scholars who devoted very considerable study to the comment, and here—both in this introduction and in the chapters that

text but who the editors also claim—usually implicitly rather than explicitly— Book Reviews follow—we go significantly beyond the standard accounts summarized so to be mistaken about their use of the Shang shu, it is startling that two scholars far (p. 4). whose first public engagement with the text seems to have come only in 2012 (just five years before the publication of this book) should question the Interestingly, among the three works considered to be the early core of scholarly integrity of senior colleagues among Western Sinologists. As for the Five Classics, the Shangshu is the only one that Western scholars the last quotation above, both the attribution of “false certainty” to others generally do not study on its own: the other two, the Classic of Poetry and the claim to uniqueness for the multi-disciplinary approach in the book (Shijing 詩經 ) and the Classic of Changes (Yijing 易經 or Zhou Yi 周 under review suggest that the editors may not know the scholarship of the field 易 ), have both received monographic studies (the Changes many more anywhere nearly as well as they suggest. Very few scholars who have studied than the Poetry); only the Shangshu, presumably because of its reputation the Shang shu have developed much sense of certainty about it. Even Wang as a mere repository of historical information, has been mostly used but Guowei 王國維 (1877–1927), almost universally regarded as the greatest of not studied. But how can one use a text one has not truly studied? One modern Chinese historians, admitted to understanding only something like half cannot, of course, and many of the flawed ways in which the Shangshu is of the text.24 And the disciplinary diversity that has been brought to the study scoured for information are precisely the result of a lack of understanding of the texts—by Chinese, Japanese, and even Western scholars—far exceeds of the nature, structure, transmission, and rhetoric of the text. Traditional assumptions about all these aspects still guide the seemingly innocuous

香港浸會大學饒宗頤國學院 use of the text as historically reliable for the era it seemingly speaks of— 23 See note 1 above. at least with regard to the chapters concerned with the Western Zhou (p. 7). 24 See the letter “Yu youren lun Shi Shu zhong shu” 與友人論詩書中成語書 (Letter with friends discussing idioms in the Poetry and the Documents) in 王國維 , In this endeavor, we have been happy to trade false certainty for more Guantang jilin 觀堂集林 , 2.1a (1923; rpt. Beijing: Zhonghua shuju, 1983), 1:75, in which he interesting and productive questions and possibilities. Ours is a collection said: “Everyone can recite the Poetry and the Documents, but they are the hardest to read of all of essays that rigorously probes the linguistic structures of individual the six classics. Given my ignorance, there’s probably half of the Documents that I am unable to explain, and even with respect to the Poetry there’s still ten or twenty percent.” 444 445

just the study of “linguistic structures” and “political ideas,” and might even be It is one of the major scholarly fallacies at the core of traditional Chinese said to go beyond the notion of “cultural memory.” philology that “early” gets equated with “reliable” (which too often then The editors seem to think that they are the first readers to have discovered inspires an ardent desire to “prove” that something is early) and that, in that the Shang shu is a collection of different types of texts by different authors turn, the demonstrable accuracy of a text is taken to prove its status not from different periods, texts moreover that have undergone a complex process only as “true” but also as a “truly early,” if not contemporaneous, witness. of transmission. [...] We are fundamentally uninterested in such “proof” because it leads only to oversimplification and reductionism (p. 8). [A]s many of our studies here reveal, a Shangshu chapter that we read today, or that scholars read as early as in Han times, was often not the I do not wish to take away anything from the contributors to this volume, result of a singular act of composition but had evolved over time, which several of whom have also adduced bronze inscriptions and “traditional turns the entire dating question into something else: not of composition Chinese philology” as evidence with which to discuss the dating of the chapters but of recomposition, compilation, and editorship, and of the dynamic under discussion. However, there are available within both the Chinese and processes of textual development over the course of the first millennium even the Western scholarly traditions more rigorous philological studies BCE. [...] to allow us to accept traditional beliefs about single authorship, regarding the Shang shu than anything on display in this volume. Despite the pristine integrity and stability of a chapter purportedly first composed this, and even despite the exaggerated claims made in the Introduction, I am in the Western Zhou, or the primacy of writing over all other forms of happy to repeat what I said at the beginning of this review: “taken as a whole textual transmission (mnemonic, performative, etc.) in early China (pp. the volume surely makes the significant contribution promised in the editors’

6–7) Introduction” and marks “a major milestone in the Western study of early Book Reviews China.” One hopes that it will be followed by more scholarship addressing To quote Captain Renault in the film Casablanca, “I am shocked, shocked” the many other chapters of the Shang shu still to be examined, or even re- to learn that the Shang shu is not marked by “pristine integrity and stability.” examinations of the chapters addressed here. However, “some scholars” (it is another disagreeable feature of the editors’ writing that over and over again they characterize—or really mischaracterize— the scholarship of anonymous others) have adduced plentiful evidence to show that the earliest texts in the Shang shu were written (and, yes, they were written) in a language that is largely comparable to that seen in Western Zhou bronze inscriptions (which, by the way, were also written). The editors seem to be particularly incensed with such evidence-based arguments, which they dismiss pejoratively as “proof” that “leads only to oversimplification and reductionism.”

In short, no matter how invested some scholars may remain in these arguments over “authenticity”—and how unshakable their faith in the 香港浸會大學饒宗頤國學院 continuous, stable written transmission of the Western Zhou chapters might be—such arguments are fundamentally bound to fail. [...] In one way or another, almost all our studies speak to questions of dating and textual transmission, but not in traditional terms. Instead, through detailed textual analysis as well as comparison with other early works, the conclusions offered in this volume are strikingly more complex (p. 6).